In a recent episode of the Omnibus Project podcast, Ken Jennings was introducing the topic of a once-popular early-twentieth century pseudo-riddle—a nonsensical question that took the form of a riddle but in reality had no answer. In order to tee up this subject, he took us on a tour through the history of riddling more generally, from ancient times to the present. Some memorable landmarks on the journey included Samson's riddle in the Bible, by which he confounds the Philistines; the riddle of the Sphinx, in the myth of Oedipus; and—in modern times—the pseudo-Medieval riddles that Gollum poses to Bilbo in The Hobbit.
Now, running down this roster, one is struck first of all by the ancient-ness and continuity of riddles: they are, as Ken tells us, the oldest known form of human puzzle-making. But one notices something else too, as soon as one inspects more closely any given one of these riddles—and that is their fundamental inadequacy. We have in our heads some notion of what a riddle should be. But every single one of the canonical examples adduced above seems to fall short of the ideal type. We can all sense that there is something just a little bit cheating or eyebrow-raising about all of them.
Take Samson's riddle, which runs—in Wikipedia's translation: "Out of the eater came something to eat, and out of the strong came something sweet." Now, Samson uses this riddle to succeed in confounding his wedding guests. They are unable to guess what Samson has in mind. But it doesn't seem like a fair contest. The solution to the riddle—a specific carcass of a lion that Samson saw, in which bees had built a honeycomb—relies upon such hyper-specific local knowledge that no one could reasonably be expected to be able to guess it, no matter how keen their intellect.
Bilbo attempts something similar with his final question to Gollum: "What's in my pocket?" But here, the denizen of the deep, Gollum né Sméagol, senses that he is cheating. It cannot be a true riddle if the answer depends on privately-held information. Yet, Gollum's own riddles seem only slightly more fair. One of his prompts runs something like this: "I am a set of white horses standing on a red hill," and the answer is supposed to be "teeth" (because they are on your gums). To which I want to respond: "Oh, okay... but you didn't tell me they could be metaphorical horses on a metaphorical hill."
The riddle of the Sphinx—though it is probably the best of the canonical examples—still suffers from the same defect. "What has four legs in the morning, two legs in the afternoon, and three legs in the evening?" only points to "man" as a solution if we accept that we weren't talking about literal legs, or literal times of day, in the first place.
We can probably all sense that there is something a little iffy about these riddles—but what exactly is wrong with them? Why is it unfair to use metaphor, etc.? Ken Jennings submits—based on the rules of the Jeopardy writers' room—that one of the key requirements in any true riddle is that the answer be "pinned." That is, the question should be framed in such a way that it could only truly lead to one solution. If it could lead to multiple answers, after all, then what makes one response more "true" or "correct" than another?
I submit that this accounts for some but not all of the defects in the canonical riddles above. The problem with the riddles that rely on metaphor, for instance, is that they are not "pinned." After all, if we allow for too wide a range of metaphorical thinking, then the solution to any riddle could be practically anything. Take Sméagol's prompt about white horses on a red hill. If we accept that by "horses," we didn't mean horses, and that by "hill," he didn't mean a hill, then why couldn't a "true" answer to the riddle be any white object on top of any red object? Or, to go even further, why couldn't "white" and "red" too have been only metaphors for other colors? Who said they literally had to be white and red?
A friend and I were trying to dissect all of this together, and we started looking up other examples of riddles to try to find out if there were any in existence that were "pinned" in the right way. One of the first examples of a riddle we came across asked about three "eyes" of different colors that signaled different instructions. The answer was supposed to be a traffic light. I was stumped by it, so my friend read the answer, but as soon as he did, I protested on the same grounds as the above: if we weren't talking about literal "eyes," then what couldn't represent eyes?
The same problem besets another example Ken Jennings gave on the podcast. It was a long riddle that had apparently never been definitively solved, although many people have claimed to have identified one possible solution as "a whale." I found this odd, because the prompt includes something about "flying through the air." But, Ken Jennings explains, proponents of the "whale" theory argue that this is metaphorical flying through metaphorical air, reflecting the way the whale swims through water. If we allow that kind of metaphorical thinking in riddles, then I suppose they're not wrong. But I also can't imagine how their solution is "pinned" enough as to be the only possible answer.
The problem of keeping an answer sufficiently "pinned" is illustrated in an episode from Harry Mathews's 1966 novel Tlooth, a notoriously unclassifiable work that combines elements of the surreal with a pastiche of 19th-century travelogues or ethnographic reports. Tlooth is itself a riddle—a hodgepodge of apparent non sequiturs combining fantastical elements in the vein of Raymond Roussel (a precursor of surrealism who influenced Mathews) that may or may not "mean" something beyond itself. A bird with teeth instead of a beak, an orchestra made of human body parts, a sun-worshiping tribe that practices human sacrifice—do they "represent" something? Or are they just... themselves?
Tlooth is such a confounding book in itself that it invites us—like the prompt of a riddle—to look for a solution. And surely one could be found. One could say the bird's teeth represent a Freudian terror of female sexuality, and on down the list. But would this interpretation be more "true" than any other? Would the solution be sufficiently "pinned" to count as the one and only correct interpretation?
Mathews teases us with the futility of trying to answer these questions by including a parable of sorts, within the text, about the art of the riddle itself. One of his characters describes an episode from their university days in which they found an inscription in a library book: "res," followed by "The Mother cannot __ her Son. The Son __ his Father..." and so on. The character proceeds from the Latin noun to the blanks to the capital letters to come up with a host of theories to explain the seemingly inscrutable words, ranging from the pornographic to the theological. Eventually, they settle on a theory grounded in Nestorian doctrine that seems to account for all the elements of the mysterious message.
The Nestorian solution appears "pinned" enough—and to incorporate all the elements in such a convincingly interlocking way—as to satisfy our demands for truth. As the narrator says, "I found a 'true' solution." They put "true" in inverted commas, however, because the solution is only "true" in the sense that it explains all aspects of the riddle. It is not "true," however, in the sense of correctly identifying the author's intent. As we learn later in the anecdote, the real reason someone wrote these words in a library book in the first place, it turns out, was not to signify the Latin noun for "thing," nor to make a theological point. Rather, they were translating from the German, retaining the German capitalization of some sentences from an exercise, and using "res" as a mnemonic to recall the spelling of the German articles.
Mathews seems to be saying something—winkingly—about his own book, and about the quest for truth more generally. He perhaps expected that his strange and sui generis novel would be subjected to many interpretations as to what it really "means." Some of them might seem "true" if they account for all the elements in the book. But would they be "true" in the sense that they reflect what Mathews meant them to mean? And if not, does that matter? Is the author's intent the criterion by which we should judge the "truth" of an interpretation?
If not author's intent, then what does make a riddle's solution true? We could say being "pinned," again, but as we see in the example of the German exercise notes, what appears at first sight to be pinned enough may actually still allow for other—more simple—solutions that we are not yet seeing. The narrator of Mathews's anecdote accepts the Nestorian solution as sufficiently pinned to be "true," but the ultimate German exercise book explanation is pinned as well, and has the advantage to recommend it as well of greater simplicity and logical parsimony. In a match up with the Nestorian theory, therefore, the German exercise theory is in something of the same relation as the Copernican to the Ptolemaic model of the solar system.
In our effort to understand what makes one riddle solution more "true" than another—in the hunt to determine when an answer is sufficiently "pinned" that it excludes all other possible answers—we have uncovered a problem with reasoning more generally. The problem of "pinning" is, in essence, the same as Quine's problem of "underdetermination." For any given facts, there will in reality be more than one available theory that could account for all of them (just as a sufficiently complicated Ptolemaic model, complete with epicycles upon epicycles, can accommodate all the facts of the solar system, as well as an elaborated Copernican model can). So why prefer one theory to another? The answer seems to have something to do with Occam's Razor—the principle of logical parsimony.
This, then, is the first task a riddle-maker faces. They must somehow minimize the problem of underdetermination (though, as Quine pointed out, it can never be wholly eliminated), by making the answer sufficiently "pinned." Where more than one "pinned" solution is possible, one answer must be preferable to the others according to Occam's Razor. The answer can't rely on so many metaphorical leaps that it is no simpler or more parsimonious than other possible solutions. It can't, for instance, depend on "flying" and "water" being mere metaphors for something other than flying and water, because at that point we have left the realm of Copernican reasoning and entered a Ptolemaic arena in which we are layering epicycles upon epicycles in order to arrive at a predetermined result: "a whale." (Using enough metaphors, after all, anything can be a whale if we really want it to be).
But as we also saw above, this is not the only challenge the riddle-maker confronts. Just as there are plenty of riddles out there that are insufficiently pinned, there is also such a thing as being pinned too much. Samson's lion, for example. Or Bilbo's pocket. There can only be one answer to what is literally inside Bilbo's pocket in the moment he asked that question. It is satisfactorily pinned. But is it so pinned—so dependent upon hyper-localized, private knowledge specific to the individual, that solving it depends simply on already knowing the answer, not on the powers of reasoning that the puzzle is supposed to test. (The same goes for Samson's riddle).
This suggests a second or third rule of riddle-making. And now that we have discovered more than one, we begin to feel the need to compile a list. It occurred to my friend and me, when we were working through the example riddles, that someone really ought to be in charge of developing such a set of binding rules for determining which riddles are fair game and which are not. There ought to be an oversight board that develops Generally Accepted Riddling Principles, or GARP, that can be used to settle disputes that arise between riddlers as to whether a given prompt is valid or not.
Such a Supreme Riddling Council does not yet exist, so this blog will have to suffice. Therefore I offer it: a list of generally accepted riddling principles that we all should agree upon. This, if you will, is the world according to GARP:
1.) A riddle should minimize the problem of underdetermination by pointing to a solution that is sufficiently "pinned";
2.) The "pinned-ness" of a riddle's solution should not depend on constructing epicycles upon epicycles; rather it should be the most logically parsimonious solution;
3.) A riddle should not depend for its solution on the use of metaphor. Each word used in the riddle should signify one or more of its commonly-accepted meanings in everyday speech;
4.) A riddle should not depend on localized private knowledge but only generally-available information for its solution.
No comments:
Post a Comment